Isiqalo esisheshayo

Yenza isicelo sokuqala: thola i-key yakho, thumela umlayezo owodwa bese ufunda impendulo. I-API ikhuluma izimo ze-OpenAI ne-Anthropic, ngakho i-SDK oyisebenzisa kakade isebenza ngemuva kokushintsha i-base URL ne-key yayo.

1. Thola i-key yakho

I-akhawunti ngayinye inayo i-API key eyodwa. Ngena, vula i-Keys & usage bese uyikopisha. I-key isebenza kuwo wonke ama-endpoint nawo wonke ama-model. Ama-key nokusetshenziswa

2. Khomba i-client yakho ku-API

Wonke ama-endpoint angaphansi kwe-base URL eyodwa. I-SDK ye-OpenAI iyamukela nge-/v1 ekugcineni; i-SDK ye-Anthropic iyamukela ngaphandle kwayo, ngoba izengeza yona i-/v1/messages.

I-client	I-base URL
Ama-SDK e-OpenAI nama-client ahambisana ne-OpenAI	`https://api.shannon-ai.com/v1`
Ama-SDK e-Anthropic	`https://api.shannon-ai.com`

Ama-endpoint amathathu akhiqiza umbhalo. Afinyelela kuma-model afanayo; khetha lowo isimo sawo ikhodi yakho esisebenzisa kakade.

I-endpoint	Isimo
`POST /v1/chat/completions`	OpenAI Chat Completions
`POST /v1/messages`	Anthropic Messages
`POST /v1/responses`	OpenAI Responses

3. Thumela isicelo sokuqala

Thumela umlayezo owodwa we-user ku-shannon-3 ku-endpoint ye-Chat Completions. Shintsha i-YOUR_API_KEY nge-key yakho.

POST https://api.shannon-ai.com/v1/chat/completions

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.shannon-ai.com/v1"
)

response = client.chat.completions.create(
    model="shannon-3",
    messages=[
        {"role": "user", "content": "Hello, Shannon!"}
    ],
    max_tokens=1024
)

print(response.choices[0].message.content)

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://api.shannon-ai.com/v1'
});

const response = await client.chat.completions.create({
  model: 'shannon-3',
  messages: [
    { role: 'user', content: 'Hello, Shannon!' }
  ],
  max_tokens: 1024
});

console.log(response.choices[0].message.content);

package main

import (
    "context"
    "fmt"
    openai "github.com/sashabaranov/go-openai"
)

func main() {
    config := openai.DefaultConfig("YOUR_API_KEY")
    config.BaseURL = "https://api.shannon-ai.com/v1"
    client := openai.NewClientWithConfig(config)

    resp, err := client.CreateChatCompletion(
        context.Background(),
        openai.ChatCompletionRequest{
            Model: "shannon-3",
            Messages: []openai.ChatCompletionMessage{
                {Role: "user", Content: "Hello, Shannon!"},
            },
            MaxTokens: 1024,
        },
    )
    if err != nil {
        panic(err)
    }
    fmt.Println(resp.Choices[0].Message.Content)
}

curl -X POST "https://api.shannon-ai.com/v1/chat/completions" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "shannon-3",
    "messages": [
      {"role": "user", "content": "Hello, Shannon!"}
    ],
    "max_tokens": 1024
  }'

I-field	Uhlobo	Okuzenzakalelayo	Incazelo
`model`	`string`		I-id ye-model ephendulayo. Ama-id abhalwe ku-Models & pricing.
`messages`	`array`		Ingxoxo kuze kube manje: ama-object ane-`role` ne-`content`. Umlayezo wokugcina we-`user` yi-prompt. Iyadingeka.
`max_tokens`	`integer`	`4096`	Isabelomali sokuphumayo sesicelo, kusuka ku-1 kuya ku-65,536. Isicelo sigcina leli nani ku-balance yakho futhi sikhokhiswa lokho esikusebenzise ngempela.
`stream`	`boolean`	`false`	I-`true` ithumela impendulo ucezu ngocezu ngenkathi ibhalwa.

Umzimba wesicelo yi-JSON futhi udinga i-header ethi Content-Type: application/json. Uhlu olugcwele lwama-field lusekhasini le-endpoint. Chat Completions

Impendulo

Isicelo esingenayo i-stream lubuyisa i-JSON object eyodwa lapho impendulo iphelele. Imibhalo nezinombolo ezingezansi yizibonelo.

200 JSON

{
  "id": "chatcmpl-9b2f6c1e4a7d4f0f8a3c5d2e1b6a7c90",
  "object": "chat.completion",
  "created": 1791590400,
  "model": "shannon-3",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! What would you like to work on?",
        "reasoning_content": "A greeting. A short, friendly reply fits."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 1386,
    "completion_tokens": 74,
    "total_tokens": 1460
  }
}

I-field	Uhlobo	Incazelo
`id`	`string`	`chatcmpl-` elandelwa yizinhlamvu ze-hexadecimal ezingu-32.
`object`	`string`	Njalo `chat.completion`.
`created`	`integer`	Isikhathi sempendulo ngamasekhondi e-Unix.
`model`	`string`	I-id eshicilelwe ye-model ephendulile, nalapho uthumele i-alias noma ukubhala okunye.
`choices`	`array`	Njalo into eyodwa, ene-`index` 0.
`choices[0].message.role`	`string`	Njalo `assistant`.
`choices[0].message.content`	`string \| null`	Impendulo. Ingaba `null` lapho impendulo iphethe i-`tool_calls`.
`choices[0].message.reasoning_content`	`string \| null`	Ukucabanga kwe-model ngaphambi kwempendulo yayo, noma `null` uma kungekho.
`choices[0].message.tool_calls`	`array`	Ikhona kuphela lapho i-model ibiza elinye lamathuluzi owachazile.
`choices[0].finish_reason`	`string`	`stop` empendulweni, `tool_calls` lapho i-model ibiza ithuluzi.
`usage.prompt_tokens`	`integer`	Ama-token i-model eyawafundile kule mpendulo. Ku-`shannon-2-*` nomndeni we-Shannon 3 lokhu kungaba ngaphezu kombhalo owuthumele.
`usage.completion_tokens`	`integer`	Ama-token i-model eyawabhala.
`usage.total_tokens`	`integer`	Isamba. Inani lokukhokha le-model ye-Shannon libalwa kulesi sibalo.

I-usage iphethe izibalo zama-token, hhayi imali. Lokho ucingo oluthathe ku-balance yakho kukhonjiswa ngesicelo ngasinye ku-Keys & usage; ukuthi kubalwa kanjani kusekhasini le-Limits and balance. Imikhawulo ne-balance
Yonke impendulo, kuhlanganise nephutha ne-stream, inayo i-header ethi x-request-id. Uma isicelo sakho siphethe i-header ethi x-request-id, inani layo liyabuya; ngaphandle kwalokho inani lizinhlamvu ze-hexadecimal ezingu-12. Ligcine kumalogi akho: lithi lesi sicelo esisodwa.

Isicelo esifanayo njenge-stream

Setha i-stream ibe true ukuze ufunde impendulo ngenkathi ibhalwa. Konke okunye esicelweni kuyahlala kunjalo.

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.shannon-ai.com/v1"
)

stream = client.chat.completions.create(
    model="shannon-3",
    messages=[
        {"role": "user", "content": "Hello, Shannon!"}
    ],
    max_tokens=1024,
    stream=True
)

for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://api.shannon-ai.com/v1'
});

const stream = await client.chat.completions.create({
  model: 'shannon-3',
  messages: [
    { role: 'user', content: 'Hello, Shannon!' }
  ],
  max_tokens: 1024,
  stream: true
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? '');
}

curl -N -X POST "https://api.shannon-ai.com/v1/chat/completions" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "shannon-3",
    "messages": [
      {"role": "user", "content": "Hello, Shannon!"}
    ],
    "max_tokens": 1024,
    "stream": true
  }'

I-frame ngayinye ye-stream ngumugqa owodwa oqala ngo-data: futhi uphethe i-JSON object. I-frame eyodwa, ephrintwe emigqeni eminingi:

I-frame eyodwa

{
  "id": "chatcmpl-9b2f6c1e4a7d4f0f8a3c5d2e1b6a7c90",
  "object": "chat.completion.chunk",
  "created": 1791590400,
  "model": "shannon-3",
  "choices": [
    {
      "index": 0,
      "delta": {
        "content": "Hello! "
      },
      "finish_reason": null
    }
  ]
}

Ama-frame, ngokulandelana	delta	finish_reason
Ngenkathi i-model icabanga. Ama-frame angu-zero noma ngaphezulu.	`{"reasoning_content": "…"}`	`null`
Impendulo, ucezu ngocezu. Ama-frame angu-zero noma ngaphezulu.	`{"content": "…"}`	`null`
I-frame yokugcina ye-JSON. Iphinde iphathe i-`usage`; akudingeki uyicele.	`{}`	`"stop"`
Ukuphela kwe-stream: umugqa othi `data: [DONE]`.

Impendulo inesimo 200 ne-content-type: text/event-stream. I-frame ngayinye ngumugqa othi data: <json> ulandelwa umugqa ongenalutho.
Umugqa othi : ping ufika njalo ngamasekhondi angu-15. Yinoma egcina uxhumano luvulekile; ama-SSE client ayiweqa.
Ukwehluleka ngemuva kokuba i-stream ivuliwe kufika njenge-frame data: {"error": {…}}. Isimo sihlala siyi-200, ngakho hlola ama-frame.

Ukusakaza

Uma isicelo sehluleka

Iphutha yi-JSON object ene-error ephethe type ne-message. Lezi yizimpendulo ezivame ukuhlangatshezwana nazo ngesicelo sokuqala.

Isimo	Uhlobo	Umlayezo	Okushoyo
`401`	`authentication_error`	`Missing authentication`	Akukho key efinyelele ku-API. Thumela `Authorization: Bearer` elandelwa yi-key yakho.
`401`	`authentication_error`	`Invalid API key`	I-key akuyona i-key yamanje ye-akhawunti. Yikopishe futhi ku-Keys & usage.
`400`	`invalid_request_error`	`unknown model: <id>`	I-`model` akuyona i-id eshicilelwe. Sebenzisa i-id evela ku-Models & pricing.
`422`	`invalid_request_error`	`Failed to deserialize the JSON body into the target type…`	I-field inohlobo lwe-JSON olungalungile, noma i-`messages` ayikho.
`429`	`rate_limit_error`	`Quota exceeded. Upgrade your plan at shannon-ai.com/plan`	Isabelomali sokuphumayo sesicelo asifaneleki ku-balance yakho.

Ukuphathwa kwamaphutha

Kuyaphi okulandelayo

Ama-model namanani — Khetha i-model: ama-id, ama-context window namanani.
Chat Completions — Wonke ama-field esicelo, ama-field empendulo nama-stream event e-endpoint oyibizile.
Ukusakaza — Ama-frame e-stream, imigqa ye-keep-alive namaphutha ngaphakathi kwe-stream.
Ukubiza umsebenzi — Vumela i-model ibize imisebenzi yakho.
Imikhawulo ne-balance — Lokho isicelo esikugcinayo nokubiza kwaso, nokuthi zingaki izicelo ongazithumela.
I-API Playground — Zama izicelo ku-browser nge-key yakho.